Oh well, spoke too soon — the Epia sounds OK for a lot of the stuff I listen to, but on more complex things like The Tallis Scholars’ Spem in Alium, or Keith Jarrett’s Hymns and Spheres, there’s a lot of ringing at the top end, probably due to the resampling the Epia does from 44.1KHz up to 48KHz, which seems to be the only datarate it will output at. Very mushy, even after the 518 has done the best it can. All right as long as I listen to Microstoria, I guess. There is talk of updated drivers which will output at 44.1, but I’m not holding my breath ;-(

A while back, I built myself a ’set top box’ based on an Epia 800 motherboard and various other bits and pieces from LinITX.com. It runs Red Hat 8.0 Linux, and I’ve used it to do a lot of investigation into video formats and playback, amongst other things. But it’s never really found a place in my life — the processor is a bit slow to use it as a compact workstation, and it won’t (quite) play back DVD-quality video. Today I has some time on my hands, so I’ve installed a wifi card, and with substantial help from the VIA Support Forums, reconfigured the audio subsystem to give me digital output (the analog out is foul). I’m using mplayer to tune into my streaming feed, and the output goes through my lovely Meridian 518 process or for dejittering, thence to my Yamaha AX-1. And the news is that the sound quality is surprisingly good — as good as I’ve ever heard 256kbs MP3s sound in fact. Good enough to actually enjoy listening to, though obvously not up to the ancient Theta Data Universal that I use for CD playback. And even better, I can control the thing over the web using a slightly-hacked version of otto from my iPaq. It’s nice when things actually Just Work.

There is a fundamental difference between the notions of place and of location. A location is completely specified by its cartography; a place owes its phenomenology to its unique context of neighbourhood and individual experience, in time. Place is what becomes of locations when things happen there.

I admit this distinction escaped me when I was first writing about augmented realities, but it has been brought home to me by projects Ben mentioned today. mudengland and mudlondon are MOOs linked to geographic metadata — very similar in concept to my original Ku24 project in Tokyo, albeit utilising much more modern technologies.

The MOOs work, but they also highlight the differences between place and location. In mudlondon, locations feel strongly quantised - to get from Oxford Circus to Hoxton you ‘go east’, but there is a lot of conceptual work to be done before that ‘going eastness’ feels different to the ‘going eastness’ which gets you from there to say Bethnal Green. By which I mean something more than it simply being, say, one move east to HMV, two to Tottenham Court Road, forty-three to Liverpool Street. The MOO is a series of linked places, but in the physical world the map is not so conveniently the territory. Place is fuzzier — real-worldlier — than a digraph of interlinked GPS coordinates. Experientially, places are contextualised by their neighbourhoods, themselves nests of neighbourhoods. Likewise, distance, intentionality and velocity alter the scale of our attention. The technical challenge is to map this experiential lustre into metadata and interfaces to humanise an otherwise arid landscape of maps.

Nice to see that the military aren’t ignoring the work on smart networks and peering. Ladies and gentlemen, DAPRA presents The Self-Healing Minefield. For the full hubris and horror, you should watch their Flash animation (follow the link). There’s something especially sickening in their representation of the mines as lurid chess knights…

We’ve been looking at satellite phones for a project in a hot place far from normal coverage. The best we’ve found so far is the Thrane & Thrane TT-3080A which has worldwide coverage through Inmarsat, and can do single-channel ISDN data transfer. The only problem, according to the very helpful sales guy at Satphone.co.uk, is that we’ll have to get our order in quickly “as we only have limited numbers of them and they are very popular with journalists going to the Middle East.”

From Associated Press:

Washington politicians would make a grave mistake in crafting new copy-protection laws based on Internet patterns today, an influential Stanford law professor warned.

The professor, Lawrence Lessig, pointed out Wednesday that millions of consumers are downloading music and other materials onto their computers because slow dial-up connections make it tough to stream content quickly to a variety of devices.

That’s bound to change within a few years as connections get faster, he said, making today’s debate irrelevant.

“In the future, it will be easier to pay for subscription services than to be an amateur database administrator who moves content from device to device,” Lessig said. “We’re legislating against a background of the Internet’s current architecture of content distribution, and this is a fundamental mistake.”

Amen to that. But I also see the long history of patinated media driving the inner compulsion towards music piracy — people are so used to having their copy of the ‘their’ music that they haven’t realised, deep inside where it matters, that ‘their’ copy is a digital clone of everyone else’s. So having a library of a gazillion downloaded MP3s feels like ‘having something of my own’ rather than ‘having access to the music content itself, whenever, wherever I am ‘. The association of music content with patina is a hard thing to break.

[Of course, there's also the fact that if it's sitting safely on your hard disk, 'they' can't take it away from you, whereas the terms and conditions on a nominal feed of content could change any day. These are paranoid times.]

When the idea of location-specificity first came up, a few years ago, I came up with a simple way to location-enable venues. The particular implementation we had in mind facilitated chance meetings of like-minded people, location-based marketing and offers flirting in social spaces. We never used it, so here it is.

• First, set up a website, where users of the system can register their contact details, mobile phone number and data about their characteristics and interests — obviously this can also be used for demographic mapping and direct marketing.
• Then, for every venue which wishes to participate, assign a unique (short) ID number. Venues can use this number in their own marketing, and more importantly, make it clearly visible at their entrances, or places within the venue where people spend some dead time — queuing at the bar, in the bacthrooms, for example.
• When people enter the venue, they see the ID for that space, and text it to a freephone number (or not free, if you want that as a revenue stream).
• From some basic characteristics of the venue, assume that once this happens, that person is likely to be there for some set period of time — maybe an hour for a bar, for example, on until that user keys in a new ID for another location. When that period elapses, their location is flagged as ‘unknown’.
• During the period that a given user is assumed to be in a given location, depending what location-specific services they’ve signed up for, they can be texted with details of the other people, offers and services available there — these could for example include ‘buy one get one free’ offers, or even the news that a single 30-something person with similar tastes just entered wearing (whatever they keyed in before leaving home).

It’s a very simple scheme, which could work well for a variety of marketing/sales purposes. I’m surprised no-one has developed this independently and tried it out.

Sky+ is fascinating — just at the point that Tivo is backing out of the UK market, Sky is working up to a major marketing push for their combined PVR/multichannel service. I think they have huge plans — if they haven’t then they’re missing a few tricks. One of my clients is pitching for the advertising business, and I’m involved in that, so I can’t post my thoughts here until the pitch is over (early March). But it’s one to think about. Interesting.

From the Wall Street Journal, via NewsScan:

A new line of wireless handsets designed to work on Wi-Fi networks will be
available in the next 12 months. The phones will be capable of handling
everything from surfing the Web to downloading music, video or large data
files such as PowerPoint presentations, and users can make voice calls to
boot. When the phones are outside of Wi-Fi coverage, users can switch over
to traditional cellular networks. Prices have not yet been set, but are
expected to cost about the same as high-end phones or PDAs, which typically
start around $500. The Wi-Fi phones likely will be bigger and heavier than
today’s diminutive wireless handsets, in order to accommodate larger
screens and possibly bigger batteries. Analysts say the advent of the new
Wi-Fi phones shows just how much plans for high-speed mobile communications
have changed over the past several years. Just a couple of years ago, the
wireless industry was touting the grand rollout of “third-generation”
wireless networks, but 3G is now late and has proven a commercial flop
where available.

Ages ago (late 90’s), we were looking at ways to facilitate adhoc working. We wanted to be able to take over a space — a café, a hotel suite, wherever, and make it instantly usable for digital collaboration. We prototyped a tool which consisted of an attache case containing a small linux box which provided file and network services, as well as routing between a dialup connection and a wireless access point (this is pre-wifi — we were using Breezecom kit). It worked and worked quite well — it was possible to plug this thing in, boot it up and everyone with wireless laptops could get on the net, share files etc. But we never really took it any further.

Then a couple of years later, wifi was standardised, Apple released the AirPort, and the whole wireless thing really took off. Strangely, I’ve never heard the AirPort discussed as a way of building adhoc working — it’s always seemed to be sold and used as either domestic access point, or for commercial infrastructure. But of course it’s still possible to put one in your bag, take it somewhere, plug it into a phone line, and open up a temporary access zone.

But AirPort and its ilk only provide access — content must be independently brought into range to be available, or fed over the Net from elsewhere. Nothing wrong with those scenarios, but I’m still interested in the potential of the emergent PacketPCs (so-called, such as the Sony FSV-PGX1 and Toshiba Hopbit). Seems a shame that these units are simply file servers, rather than a programmable platform (although I’d be surprised if the hackers don’t get onto that issue fairly quickly!). But even so, they’re interesting. Is PacketPC the concept for the future for adhocery? The current generation still don’t seem to be the right thing to carry around always on (I wonder about the battery life, for instance), but they push in the right direction And at least they make the idea of the adhoc collaborative space more attractive than an attache-case full of hacked laptop and wireless kit!